Tripath TA2024/20 Bridge/Parallel Config

[Pano]

Intersting board, John. What is it out of? A HT in a Box?

[john65b]

Aloha Panomaniiac...

Nope, they came out of Video Game Kiosk. Actually, they were R&D fabs. Don't know if it ever flew or not. Got a bunch of them and a few 6 ch boards have inoperable channel/channels.

Put up a few for sale on Audiogon.

Preliminary tests confirm they put out more power than the TA2024 (duh), but lack some of the magic of the TA2024. We are still putzing with some of the power supplies (12 V 7Ah battery currently) and components (input/output caps to start) to see if we can regain that better soundstage and imaging from the TA2024.

From what I read, you have some profound knowledge/experience with these boards, any pointers? The SMD components on these bords seem to be laid out exactly like the schematic for the Tripath TA2020 datesheet, except they seem to have component values 10x bigger (ie 10 mH and the board has 100 mH inductors) - maybe 'cause of the six outputs sucking off the same power rails?

Anyway, your input is welcomed.

Rich - are you getting this?

BTW, transferred to Kapolei on Oahu in 1999 for work and lived there for 3.5 years... currently live in Utah...miss the beaches, Girls in skimpy bikinis, SCUBA Dives, Girls in skimpy bikinis, Kona beer and Volleyball on the beach, and yes, Spam Misubies (mainlanders will never understand the innate Hawaiian love of Spam - yum). Did I mention Girls in skimpy bikinis?

Mahalo

[BWRX]

Originally posted by john65b
BWRX --> You said four resistors in your previous post, and from the LM3886 datahseet outlining the parallel mode, you only need to put the four resistors on the four positive outputs and the neg outputs go to ground resistorless ...

I said 4 because I thought you'd only be testing one channel instead of a stereo setup. The LM3886 only need resistors on one output because they have single ended outputs. These Tripath chips need resistors on both outputs because they're bridged - the negative output is not grounded. This is just another example of how much punishment the chips can take!

[john65b]

Whoops, my bad.

I will re-try tonight. I can always put these resistors at the speaker to make the wiring eaiser - no soldering at the board outputs, but double the wires.

If this arrangement works and sounds good, I will investigate paralleling my XR50 Panny outputs (party mode and reversed channels) to my Maggies to see if I can get 175wpc of pure digitally amplified love to them.

Whooo hooo...

[john65b]

After having all eight resistors on the outputs, the sound was a bit better, but I firmly believe my linear power supply is insufficient for these four channels.

I have two TA2024 amps and I will try them tomorrow, since power requirements would not be an issue here...

So far, so good.

[Pano]

Originally posted by john65b
From what I read, you have some profound knowledge/experience with these board

Uh, well - yeah, I guess I do.
Everyone needs a hobby, right?

they seem to have component values 10x bigger (ie 10 mH and the board has 100 mH inductors)

If they do, there is something very wrong. That would bring the filter -3dB point down by 10X, not a good thing. BTW, it's 10uH MICRO Henries, not milli. That would make a huge difference, too!

Yep, that's Hawaii, girls in bikinis, spam and coconuts. Gotta love it, brah!

[john65b]

As some else also replied, your must likely correct, these, the component values have to be correct...I must be getting my mili/micro mixed up...

Briefly tried the two Sonic Impact TA2024 boards paralleled, same difference. muddy, and incoherent. I may try again with a 1 ohm resistor on each output to see if that makes any diference...

more later